about Epidemiology & the department

Epidemiology academic information

Epidemiology faculty

Epidemilogy resources

sites of interest to Epidemiology professionals

Frerichs, R.R. Study on Breastfeeding and HIV. 

SEA-AIDS Network, August 27, 1999.

R.R. Frerichs Posting

The breastfeeding debate has captured the attention of SEA-AIDS members, similar to other groups in developing countries. The issue is, should breastfeeding be discouraged among HIV infected women? While mother's milk is important for the development of the offspring, as has been strongly encouraged by agencies and experts alike, feeding milk contaminated with virus has a counter price that all too often is met with death. But how often? This point is addressed in an article published in the current Journal of the American Medical Association (Vol. 282, pp. 744-749, August 25, 1999). 

To address the breastfeeding issue, decision-makers need to know what percentage of infants of HIV infected women become infected by breastmilk, separate from infection coming from the blood of the mother during the intrauterine or birth period. The authors of the JAMA study were able to make such a determination. They did a prospective cohort study (i.e., followed a group forward in time) in Malawi, Africa of 672 infants. This specially-selected group was born to HIV infected women, but was free of apparent HIV during the first weeks of life (i.e., all were at risk when breastfeeding got underway). None of the women in this group had been treated with retroviral drugs during or after pregnancy. After one month of being breast fed, 3.5% of the infants were infected due to breastmilk. The cumulative percentage of being infected due to breastmilk rose to 7.0% at the end of 11 months, and finally to 10.3% after 23 months of breastfeeding. These estimates are lower than the full impact of breastfeeding since infants who were infected during the first few weeks of life were excluded from the cohort. Nevertheless, they provide a clear picture of the rising risk with time, as the duration of breastfeeding continues.

The Malawi results provide vital data for public health and medical decision-makers and add to our current knowledge. Several years earlier, investigators had summarized the findings of five studies in a meta-analysis. They estimated that 14% of births to HIV infected women would become infected through breastfeeding (The Lancet 1992;340585-588).

Missing at the local level throughout the developing world is information on what happens to births of HIV infected women who are not breast fed. Some will receive supplemental milk, as is done in Thailand. Others will try sugar water or similar poor solutions, and face the risk of death from enteric pathogens. Here more assistance is needed from SEA-AIDS members on low-cost, but healthy, alternatives to breastfeeding for HIV infected women in Asia. What has been tried in various countries of the region? What level of infection or infant death has been associated with the alternative feeding schemes? How do these negative health outcomes compare to the "HIV- due-to-breastfeeding" percentages reported by the Malawi study? 

Once such answers are known, appropriate solutions will be easier to derive.